p53 - - PDF document

大腸直腸腺瘤及腺癌之 p53 蛋白表現在台灣：兩家醫院之研究 曾志恩 陳榮達 1 何 霖 1

摘要 目的：1.探討大腸直腸腺瘤及腺癌之 p53 蛋白表現在台灣的現狀及與其他族群是 否不同？2.p53 蛋白可否當做大腸纖維鏡術的一種輔助性檢查？材料與方法：收 集福馬林固定處理後之腺瘤 110 個，源發自腺瘤之腺癌 46 個及腺癌 62 個，進行 常規 p53 蛋白之免疫組織化學染色並統計分析其結果。結果：腺瘤、源發自腺瘤 之腺癌及腺癌之平均年齡分別為 63.3 歲、67.1 歲和 69.3 歲；其 p53 蛋白陽性率分 別為 6.4%，28.3%/32.6%和 71.0%。腺瘤之纖毛型態及異生嚴重度與 p53 蛋白表現 無正相關。結論：1.大腸直腸腺瘤及腺癌之 p53 蛋白表現在台灣與其他族群並無 不同。2.腺瘤的絨毛和異生嚴重度與 p53 蛋白表現不相稱。3.從腺瘤至腺癌，p53 蛋白表現逐漸增加，支持前人腺瘤－腺癌系列之假說。從腺瘤至長成腺癌約需 6 年。4.對低度異生或絨毛組成較少但 p53 蛋白過度表現之切除不完全腺瘤，應積 極的進一步完全切除。(慈濟醫學 2003; 15:305-310) 關鍵語：p53 蛋白，大腸直腸，腺瘤，腺癌 P53 Protein Presentation of Colorectal Adenomas and Carcinomas in Taiwan: Experience

• f Two Hospitals

Jeh-En Tzeng, Jung-Ta Chen1, William L. Ho1 Department of Pathology, Buddhist Dalin Tzu Chi General Hospital, Chiayi, Taiwan; Department of Pathology1,Taichung Veterans General Hospital, Taichung, Taiwan

ABSTRACT Objective: We first attempted to determine if the p53 protein presentation of colorectal adenomas and carcinomas showed any difference between Taiwanese and other

• populations. Second, as incompletely polypectomized or biopsied adenomas of the

colorectum are a troublesome issue in routine practice, we sought to determine if p53 protein is a helpful indicator for clinicians in deciding the subsequent therapeutic mode for incompletely removed tumors. Materials and Methods: We studied formalin-fixed tissue, including 110 adenomas, 46 adenocarcinomas arising from adenomas, and 62 frank adenocarcinomas with routine immunohistochemical staining and statistical analysis. Results: The average ages of patients with adenomas, adenocarcinomas arising from adenomas, and frank adenocarcinomas were 63.3, 67.1, and 69.3 years. The p53 protein-positive rates of these 3 groups were 6.4%, 28.3%/32.6% (adenomatous part/adenocarcinomatous part), and 71.0%, respectively. Conclusions: First, there was no statistical difference in colorectal p53 protein presentation between Taiwanese and other

• populations. Second, histologically, the villous morphology and dysplastic severity of the

adenoma were not correlated with p53 protein expression. Third, the sequential increase in p53 protein overexpression from adenomas to frank adenocarcinomas supports the hypothesis of an adenoma-carcinoma sequence in polypoid colorectal tumors. The transformation duration from adenomas to carcinomas is about 6 years. Fourth, an incompletely polypectomized adenoma, which shows low dysplasia or a small villous component but p53 protein overexpression, should be aggressively completely resected. (Tzu Chi Med J 2003; 15:305-310) Key words: p53 protein, colorectum, adenoma, adenocarcinoma Received: April 3, 2003, Revised: May 19, 2003, Accepted: June 16, 2003

Address reprint requests and correspondence to: Dr. Jeh-En Tzeng, Department of Pathology, Buddhist Dalin Tzu Chi General Hospital, 2, Min Sheng Road, Dalin, Chiayi, Taiwan INTRODUCTION Colorectal cancer is a globally important cause of cancer-related death. Abnormal cell growth and differ-entiation, associated with the accumulation of genetic alternations over a long time, are the etiology of colo-rectal cancer, and a genetic model for colorectal tumorigenesis is well established [1]. The p53 gene is located on the short arm of chromosome 17, which encodes a 53-kD phosphoprotein. This protein might play a certain role in the regulatory control of normal cell prolif-eration. The wild-type p53 protein usually does not accumulate in amounts detectable by immunohistochemistry because of its short half-life of 6-20 minutes. However, the mutant type has a half-life of up to 6 hours [2]. This functionally inactive and stabilized p53 protein can be detected in nuclei of cells. By using this nature of the p53 protein, many researchers focusing on the relationship of p53 protein or the p53 gene with colorectal cancer have obtained significant results [3-11]. But, there is still limited study of p53 in Taiwan to our knowledge. The first aim of this study was to demonstrate colorectal p53 protein expression in Taiwan and to determine if any differences exist with other popula-tions. Clinically, an incompletely polypectomized ade-noma or incompletely biopsied adenoma is

• ccasionally encountered when colonoscopy is performed. In those situations, what is the

best suggestion for clinicians? An indicator, which could help distinguish high-risk adenomas and which is easily detected, is necessary. Now that the p53 gene is known to play a key role in the adenoma-carcinoma sequence [1], it seems reasonable to use the p53

protein as an indicator for this purpose. The secondary purpose of this study was to determine whether the p53 protein, detected by an immunohistochemical method, is a good adjuvant indicator for colonoscopy. MATERIALS AND METHODS Archival formalin-fixed, paraffin wax-embedded tissues from the Buddhist Dalin Tzu Chi General Hospital and Taichung Veterans General Hospital of Taiwan were used. During August 1999 to February 2002, 110 adenomas from 89 patients were studied after colofiberoscopic biopsies or polypectomies. Dysplasia of the adenomas was graded on a 2-tier scale as either mild/moderate or severe as per the WHO classification. The villous morphology of the adenoma was classified into 3 groups: less than 25%, between 25% and 75%, and more than 75% of the mucosal surface. Sixty-two frank adenocarcinomas, which were defined as adenocarcinomas with tumors invading beyond the mucosa muscularis of the colorectal wall, from 60 patients including 13 cancers via colonoscopic biopsies and 49 via excisions, were collected. A single block representative of the tumor was selected in each case. Forty-six adenocarcinomas arising from adenomas including surgically removed specimens and colonoscopic biopsies were obtained from 45 patients. For the immunohistochemical staining of p53 protein, formalin-fixed, parafin-embedded tissue was used. Three-micrometer-thick sections were deparafin-ized in xylene, re-hydrated in a series of graded alcohol, and later exposed to 3% hydrogen peroxidase to drive off the endogenous peroxidase. Sections were immersed in cuvettes with citrate buffer (pH 6.0), which were placed in a bowl containing 1500 mL tap water. The bowl was transferred into a microwave oven (National, Tou-Yan, Taiwan) with an automatic rotating plate, and irradiated at 720 W for 10 minutes for antigen retrieval [12]. After microwave

treatment, sections were pre-incubated in PBS. Monoclonal antibody DO-7 (Dako, Denmark, code M7001), which reacts with wild-type and mutant-type human p53 protein, was applied to the sections (dilution 1:50; incubation 30 minutes). Biotinylated rabbit anti-mouse antibody (Dako, code K0672 bottle 2; incubation 10 minutes) was used as the secondary antibody. The immunoreaction was visualized using the avidin/biotin complex (Dako, code K0672 bottle 3; incubation 10 minutes) with hydrogen peroxide as the substrate and N,N-dimethylformamide (DMF) as a chromogen. Counterstaining was performed with Mayer's hematoxylin. Positive control was performed by using colonic carcinomatous sections with high p53 overexpression. Normal mucosa was stained as a negative control. Sections from each tumor were examined at X40 to X400 magnification. The ratio of the p53 protein-positive nuclei was blindly calculated by 2 pathologists. The presence of more than 10% p53 protein-positive nuclei was regarded as being a p53 protein-positive presenta-tion, and a consensus was reached when a discrepancy was encountered between the 2 pathologists. p53 protein-positive nuclei were visualized as having dispersed or compact patterns [13]. The former is spottily distributed and usually only involves a small proportion of glandular nuclei. The latter shows a contiguous area of p53-reactive nuclei and usually involves more-glandular nuclei. RESULTS Seven adenomas (6.4%) of these 110 adenomas had evidence of p53 protein

• verexpression. The mean age of the 89 cases (54 males and 35 females) was 63.3 years

(Table 1). The p53 protein reaction was always localized in the nuclei, and all positive cases had the compacted pattern (Fig. 1). Twelve (11.7%) of the 103 negative adenomas

had less than 10% p53 protein-positive nuclei but with the compacted pattern. Fourteen of the 110 adenomas had less than 25% villous part, 82 ade-nomas had 25%-75% villous part, and 14 adenomas had more than 75% villous part. Ninety-five of the 110 ade-nomas had mild to moderate dysplasia, and 15 adenomas had severe dysplasia. All 7 p53 positive-cases had 25%-75% villous parts and showed mild to moderate dysplasia. Neither the amount of the villous part nor the severity of dysplasia was correlated to p53 protein

• verexpression.

Sixty patients with 62 frank adenocarcinomas, which included 13 cancers retrived via colonoscopic biopsies and 49 via excisions, were studied. Forty- four adenocarcinomas (71.0%), which included 8 (61.5%) of the 13 biopsied cancers and 36 (73.5%) of the 49 excised cancers, had p53 protein overexpression. The mean age of these 60 cases (39 males and 21 females) was 69.3 years (Table 1). All p53-positive cases showed the compacted pattern and only 1 of the negative adenocarcinomas focally demonstrated the compacted

• pattern. Another case of a negative adenocarcinoma showed a uniform dispersed pattern.

In the group of 46 adenocarcinomas arising from adenomas, the mean age of the 45 cases (29 males and 16 females) was 67.1 years (Table 1). Eleven (23.9%) of the 46 tumors had p53 protein overexpression in both the adenomatous parts and cancerous parts (Fig. 2). Twenty-nine (63.0%) of the 46 tumors showed no p53 reaction in either the adenomatous part or the cancerous part. Four tumors showed a positive p53 reaction in the cancerous part, but were negative in the adenomatous part. The last 2 tumors demonstrated a positive p53 reaction in the adenomatous part, but were negative in the cancerous part (Table 2). DISCUSSION

Nearly all p53-positive stains were presented within nuclei. The cytoplasma either negatively stained in most cases or very weakly stained in some cases in our study. Nuclear accumulation of the p53 protein, as previously reported [13], was visualized as following 2 distinct patterns: compact or dispersed. These 2 distinct patterns were correlated to the rate of positive nuclei. When the rate of nuclear staining was greater than 10% of tumor nuclei, the majority of positively stained nuclei had the compact pattern. However, when the p53 protein-positive rate was less than 10% of tumor nuclei, positively stained nuclei always showed the dispersed pattern. The correlation between positively stained nuclear rates and their distributive patterns indicated that the 10% cut-off value for p53-positive staining is reasonable. Twelve cases (11.6%) of adenomas showed less than a 10% p53 protein-positive presentation but had the focally compact positive p53 nuclear pattern. This phenomenon might have been due to focally reactive changes or an early sequential transformation from an adenoma to an invasive carcinoma [13]. Seven cases (6.4%) of adenomas had positive p53 expression. This finding was similar to results of Kikuchi et al [14] and Van et al [15], both of which showed 8% with positive p53 expression. Histologically, neither the villous ratio nor the severity of dysplasia of the adenomas was correlated to p53 protein accumulation. These findings are similar to those

• f previous reports [3,11,16,17], but differ from Kikuchi et al.'s results [14]. The latter

showed an increasing p53 mutation rate of from 0%, 8%, 15%, to 40% in adenomas with moderate dys-plasia, adenomas with severe dysplasia, intramucosal carcinomas, and invasive carcinomas, respectively. The reason why we failed to demonstrate a relationship between p53 expression and the degree of dysplasia is unclear. But, it might have been due to a focally small number of p53-positive glands not being found because of a lack of

series cuts to the entire adenoma [16] or to missense mutations at the genetic level [18]. Forty-four tumors (71%) of 62 frank adenocarcinomas showed p53 overexpression. There was no obvious difference in p53 protein overexpression (p > 0.05) between biopsied specimens (8/13; 61.5%) and excised specimens (36/49; 73.5%). This ratio was similar to well-documented reports of p53 protein [3,5,9,11,19,20] and gene [21,22] presentations of colorectal adenocarcino-mas. Those studies showed the p53 protein positive rate to be within 59.6% to 72%, and, the rate of the p53 genetic abnormality to be 60% to 75% in colorectal cancers. This finding, in our study, demonstrates that p53 protein overexpression in Taiwanese is similar to that seen in other populations. In the group of adenocarcinomas arising from ade-nomas (Table 2), the p53 protein-positive rate of the adenomatous part was 28.3% (13/46) and of the adenocarcinomatous part was 32.6% (15/46). There was no statistical difference in p53 protein presentation between the adenomatous part and the adenocarinomatous part in this group (p > 0.05). This group's result was between the pure adenomatous group, which showed a 6.4% p53 protein-positive rate, and the frank adenocar-cinomatous group, which showed a 71.0% p53 protein-positive rate, in this study. Obviously, there was a sequential increase in p53 protein overexpression from adenomas to adenocarcinomas arising from adenomas, and then to frank adenocarcinomas. This result is similar to previous reports of Auvinen et al [5] and Vogel-stein et al [21], and supports the hypothesis of the ade-enoma-carcinoma sequence. It was interesting to find out its next change in a p53 protein-positive adenoma. At the genetic level, these p53-positive adenomas had either a point mutation or allelic deletion [2,7,14,19,21-24], and these changes were irreversible. Due to loss of normal regulatory function of cellular proliferation of the defective p53 gene, it is reasonable to predict that the p53 protein-positive focus of an adenoma would expend and transform to a malignancy. This transformation process had no return, and the p53

protein showed sequentially increasing accumulation. This phenomenon makes the p53 protein an adjuvant indicator for clinicians to decide on a more-aggressive therapeutic policy in an incompletely polypectomized or biopsied p53-positive adenoma. The 2 adenocarcinomas arising from adenomas which were p53-postive in the adenomatous parts but p53-negative in the adenocarcinomatous parts had smaller malignant foci within relatively larger adenomas. About 5% of the nuclei of these 2 malignant foci stained p53-postive, but their adenomatous counterparts had more than 10% p53-positive nuclei. The exact reason for this phenomenon is unclear. But, it might have been due to a bias of small foci with early malignant changes or another p53 gene mutation coding for a stop code and inducing no products of p53 protein during the cancerous transformation. The mean ages of these 3 groups also indicated sequential changes: 63.3, 67.1, and 69.3 years, respectively. Only the age difference between the adenomatous group and the frankly adenocarcinomatous group reached statistical significance (p < 0.05). However, this finding roughly predicts the transformation from adenomas to frank adenocarcinomas to be about 6 years and supports Muller et al's study [25] which showed that a colonoscopy maintained a protective influence for 6 years. CONCLUSIONS Colorectal p53 protein overexpression in Taiwanese is similar to those of related reports for other populations. Histologically, the villous morphology and dysplastic severity of the adenoma were not correlated with p53 overexpression. The sequential increase in p53 protein overexpression from adenomas to frank carcinomas supports the previous

hypothesis of an adeno-ma-carcinoma sequence in polypoid colorectal tumors. An incompletely polypectomized adenoma, which shows low dysplasia or a small villous component but p53 protein overexpression, should be aggressively completely resected. REFERENCES

• 1. Fearon ER, Vogelstein B: A genetic model for colorectal tumorigenesis. Cell 1990;

61:759-767.

• 2. Levine AJ, Momand J, Finlay CA: The p53 tumour suppressor gene. Nature 1991;

351:453-456.

• 3. Yamaguchi A, Nakagawara G, Kurosaka Y, Nishimura G, Yonemura Y, Miyazaki I:

P53 immunoreaction in endoscopic biopsy specimens of colorectal cancer, and its prognostic significance. Br J Cancer 1993; 68:399-402.

• 4. Leahy DT, Salman R, Mulcahy H, Sheahan K,O'Donoghue DP, Parfrey NA: Prognostic

significance of p53 abnormalities in colorectal carcinoma detected by PCR-SSCP and immunohistochemical analysis. J Pathol 1996; 180:364-370.

• 5. Auvinen A, Isola J, Visakorpi T, Koivula T, Virtanen S, Hakama M: Overexpression of

p53 and long-term survival in colon carcinoma. Br J Cancer 1994; 70:293-296.

• 6. Bosari S, Viale G, Bossi P, et al: Cytoplasmic accumulation of p53 protein: An

independent prognostic indicator in colorectal adenocarcinomas. J Natl Cancer Inst 1994; 86:681-687.

• 7. Caldes T, Iniesta P, Vega FJ, et al: Comparative survival analysis of p53 gene mutations

and protein accumulation in colorectal cancer. Oncology 1998; 55:249-257.

• 8. Scott N, Sagar P, Stewart J, Blair GE, Dixon MF, Quirke P: P53 in colorectal cancer:

Clinicopathological correlation and prognostic significance. Br J Cancer 1991; 63: 317-319.

• 9. Mulder JW, Baas IO, Polak MM, Goodman SN, Offerhaus GJA: Evaluation of p53

protein expression as a marker for long-term prognosis in colorectal car-cinoma. Br J Cancer 1995; 71:1257-1262.

• 10. Zeng ZS, Sarkis AS, Zhang ZF, et al: P53 nuclear overexpression: An independent

predictor of survival in lymph node-positive colorectal cancer patients. J Clin Oncol 1994; 12:2043-2050.

• 11. Yamaguchi A, Kurosaka Y, Fushida S, et al: Expression of p53 protein in colorectal

cancer and its relationship to short-term prognosis. Cancer 1992; 70:2778-2784.

• 12. Hedrum A, Ponten F, Ren Z, Lundeberg J, Ponten J, Uhlen M: Sequence-based

analysis of the human p53 gene based on microdissection of tumor biopsy samples. Bio Techniques 1994; 17:118-129.

• 13. Ren ZP, Ponten F, Nister M, Ponten J: Two distinct p53 immunohistochemical patterns

in human squamous-cell skin cancer, precursors and normal epidermis. Int J Cancer 1996; 69:174-179.

• 14. Kikuchi-Yanoshita R, Konishi M, et al: Genetic change of both p53 alleles associated

with the conversion from colorectal adenoma to early carcinoma in familial adenomatous polyposis and non-familial adenomatous polyposis patients. Cancer Res1992; 52: 3965-3971.

• 15. Van den Berg FM, Tigges AJ, Schipper MEI, Den Hartog-Jager FCA, Kroes WGM,

Walboomers JMM: Expression of the nuclear oncogene p53 in colon tumors. J Pathol 1989; 157:193-199.

• 16. Kawasaki Y, Monden T, Morimoto H, et al: Immunohistochemical study of p53

expression in microwave-fixed, paraffin-embedded sections of colorectal carcinoma and

• adenoma. Am J Clin Pathol 1992; 97:244-249.
• 17. Porter PL, Gown AM, Kramp SG, Coltrera MD: Widespread p53 overexpression in

human malignant tumors. An immunohistochemical study using methacan-fixed, embedded tissue. Am J Pathol 1992; 140:145-153.

• 18. Dix B, Robbins P, Carrello S, House A, Iacopetta B: Comparison of p53 gene mutation

and protein overexpression in colorectal carcinomas. Br J Cancer 1994; 70:585-590.

• 19. Campo E, de la Calle-Martin O, Miquel R, et al: Loss of heterozygosity of p53 gene

and p53 protein expression in human colorectal carcinomas. Cancer Res 1991; 51: 4436-4442.

• 20. Lanza G Jr, Maestri I, Dubini A, et al: P53 expression in colorectal cancer: Relation to

tumor type, DNA ploidy pattern and short-term survival. Am J Clin Pathol 1996; 105:604-612.

• 21. Vogelstein B, Fearon ER, Hamilton SR, et al: Genetic alterations during

colorectal-tumor development. N Engl J Med 1988; 319:525-532.

• 22. Cripps K J, Purdie CA, Carder PJ, et al: A study of stabilisation of p53 protein versus

point mutation in colorectal carcinoma. Oncogene 1994; 9:2739-2743.

• 23. Hollstein M, Sidransky D, Vogelstein B, Harris CC: P53 mutations in human cancers.

Science 1991; 253:49-53.

• 24. Baker SJ, Preisinger AC, Jessup JM, et al: P53 gene mutations occur in combination

with 17p allelic deletions as late events in colorectal tumorigenesis. Cancer Res 1990; 50:7717-7722.

• 25. Muller AD, Sonnenberg A: Prevention of colorectal cancer by flexible endoscopy and
• polypectomy. A case-control study of 32,702 veterans. Ann Intern Med 1995;123:904-910.

• Fig. 1. A p53 protein-positive adenoma showing p53-reactive nuclei and the

compacted nuclear pattern (right); a p53 protein-negative adenoma demonstrating the dispersed pattern of scattered spotty p53-reactive nuclei (left). (200X)

• Fig. 2. Adenocarcinoma arising from an adenoma showing positive p53 reactivity in

both (right) the adenocar-cinomatous part and (left) the adenomatous part. (400X) Table 1. Basic Data and p53 Protein Presentations of Adenomas, Adenocarcinomas Arising from Adenomas and Frank Adenocarcinomas

Adenomas #Ca from adenomas Adenocarcinomas

• No. of tumor

110 46 62

• No. of cases

89 45 60 M:F 54:35 29:16 39:21 Mean age (years) 63.3 67.1 69.3 p53+ 6.4%(7/110) *28.3%/32.6% 71.0%(44/62) #: Ca from adenomas: Adenocarcinomas arising from adenomas; *: Adenomatous part/adenocarcinomatous part Table 2. Immunohistochemical Presentation of p53 Protein in 46 Colorectal Adenocarcinomas Arising from Adenomas Acp Ap p53 positive p53 negative Total p53 positive 11 (23.9%) 2 ( 4.4%) 13 (28.3%) p53 negative 4 ( 8.7%) 29 (63.0%) 33 (71.7%) Total 15 (32.6%) 31 (67.4%) 46 (100%) Acp: adenocarcinomatous part; Ap: adenomatous part